Auto-evaluation of skull radiograph accuracy using unsupervised anomaly detection.

BACKGROUND: Radiography plays an important role in medical care, and accurate positioning is essential for providing optimal quality images. Radiographs with insufficient diagnostic value are rejected, and retakes are required. However, determining the suitability of retaking radiographs is a qualitative evaluation. OBJECTIVE: To evaluate skull radiograph accuracy automatically using an unsupervised learning-based autoencoder (AE) and a variational autoencoder (VAE). In this study, we eliminated visual qualitative evaluation and used unsupervised learning to identify skull radiography retakes from the quantitative evaluation. METHODS: Five skull phantoms were imaged on radiographs, and 1,680 images were acquired. These images correspond to two categories: normal images captured at appropriate positions and images captured at inappropriate positions. This study verified the discriminatory ability of skull radiographs using anomaly detection methods. RESULTS: The areas under the curves for AE and VAE were 0.7060 and 0.6707, respectively, in receiver operating characteristic analysis. Our proposed method showed a higher discrimination ability than those of previous studies which had an accuracy of 52%. CONCLUSIONS: Our findings suggest that the proposed method has high classification accuracy in determining the suitability of retaking skull radiographs. Automation of optimal image consideration, whether or not to retake radiographs, contributes to improving operational efficiency in busy X-ray imaging operations.

Comparison of qualitative and fully automated quantitative tools for classifying severity of white matter hyperintensity.

OBJECTIVE: In this study, we aimed to compare the Fazekas scoring system and quantitative white matter hyperintensity volume in the classification of white matter hyperintensity severity using a fully automated analysis software to investigate the reliability of quantitative evaluation. MATERIALS AND METHODS: Patients with suspected cognitive impairment who underwent medical examinations at our institution between January 2010 and May 2021 were retrospectively examined. White matter hyperintensity volumes were analyzed using fully automated analysis software and Fazekas scoring (scores 0-3). Using one-way analysis of variance, white matter hyperintensity volume differences across Fazekas scores were assessed. We employed post-hoc pairwise comparisons to compare the differences in the mean white matter hyperintensity volume between each Fazekas score. Spearman's rank correlation test was used to investigate the association between Fazekas score and white matter hyperintensity volume. RESULTS: Among the 839 patients included in this study, Fazekas scores 0, 1, 2, and 3 were assigned to 68, 198, 217, and 356 patients, respectively. White matter hyperintensity volumes significantly differed according to Fazekas score (F=623.5, p<0.001). Post-hoc pairwise comparisons revealed significant differences in mean white matter hyperintensity volume between all Fazekas scores (p<0.05). We observed a significantly positive correlation between the Fazekas scores and white matter hyperintensity volume (R=0.823, p<0.01). CONCLUSIONS: Quantitative white matter hyperintensity volume and the Fazekas scores are highly correlated and may be used as indicators of white matter hyperintensity severity. In addition, quantitative analysis may be more effective in classifying advanced white matter hyperintensity lesions than the Fazekas classification.

Automated angular measurement for puncture angle using a computer-aided method in ultrasound-guided peripheral insertion.

Ultrasound guidance has become the gold standard for obtaining vascular access. Angle information, which indicates the entry angle of the needle into the vein, is required to ensure puncture success. Although various image processing-based methods, such as deep learning, have recently been applied to improve needle visibility, these methods have limitations, in that the puncture angle to the target organ is not measured. We aim to detect the target vessel and puncture needle and to derive the puncture angle by combining deep learning and conventional image processing methods such as the Hough transform. Median cubital vein US images were obtained from 20 healthy volunteers, and images of simulated blood vessels and needles were obtained during the puncture of a simulated blood vessel in four phantoms. The U-Net architecture was used to segment images of blood vessels and needles, and various image processing methods were employed to automatically measure angles. The experimental results indicated that the mean dice coefficients of median cubital veins, simulated blood vessels, and needles were 0.826, 0.931, and 0.773, respectively. The quantitative results of angular measurement showed good agreement between the expert and automatic measurements of the puncture angle with 0.847 correlations. Our findings indicate that the proposed method achieves extremely high segmentation accuracy and automated angular measurements. The proposed method reduces the variability and time required in manual angle measurements and presents the possibility where the operator can concentrate on delicate techniques related to the direction of the needle.

Evaluation of deep learning reconstruction on diffusion-weighted imaging quality and apparent diffusion coefficient using an ice-water phantom.

This study assessed the influence of deep learning reconstruction (DLR) on the quality of diffusion-weighted images (DWI) and apparent diffusion coefficient (ADC) using an ice-water phantom. An ice-water phantom with known diffusion properties (true ADC = 1.1 × 10-3 mm2/s at 0 °C) was imaged at various b-values (0, 1000, 2000, and 4000 s/mm2) using a 3 T magnetic resonance imaging scanner with slice thicknesses of 1.5 and 3.0 mm. All DWIs were reconstructed with or without DLR. ADC maps were generated using combinations of b-values 0 and 1000, 0 and 2000, and 0 and 4000 s/mm2. Based on the quantitative imaging biomarker alliance profile, the signal-to-noise ratio (SNRs) in DWIs was calculated, and the accuracy, precision, and within-subject parameter variance (wCV) of the ADCs were evaluated. DLR improved the SNR in DWIs with b-values ranging from 0 to 2000s/mm2; however, its effectiveness was diminished at 4000 s/mm2. There was no noticeable difference in the ADCs of images generated with or without implementing DLR. For a slice thickness of 1.5 mm and combined b-values of 0 and 4000 s/mm2, the ADC values were 0.97 × 10-3and 0.98 × 10-3mm2/s with and without DLR, respectively, both being lower than the true ADC value. Furthermore, DLR enhanced the precision and wCV of the ADC measurements. DLR can enhance the SNR, repeatability, and precision of ADC measurements; however, it does not improve their accuracies.

[Evaluation of Detectability in Clinical Practice of Metal Detectors, etc. for Magnetic Resonance Imaging].

PURPOSE: In order to prevent magnetic materials from being brought into the magnetic resonance imaging (MRI) examination room, many facilities have metal detectors, etc., but there are various types of equipment with different performance and characteristics. The purpose of this study was to evaluate each detector in actual clinical practice. METHODS: At multiple facilities, gate-type magnetic detectors, pole-type magnetic detectors, handy-type magnetic detectors, and handy-type metal detectors were used to identify 9 types of objects that may be brought into the MRI examination room. We performed evaluation of detection distance measurement assuming actual operation. RESULTS: The gate type was only able to detect objects with strong magnetism. With the pole type, the closer the measurement distance was to the pole, the more objects could be detected, and the lower the pole, the shorter the detection distance. With the handy type, there were many objects that could be detected when the device and the object were brought into close contact. CONCLUSION: The detectability of the instruments varied depending on the size and type of the object. It is important to understand the characteristics of each device and use it according to the purpose in carrying-in confirmation before the examination.

[Echo Train Length (ETL) of Fluid-attenuated Inversion Recovery (FLAIR) and Extraction Volume of White Matter Hyperintensity Volume in Automated White Matter Signal Analysis].

PURPOSE: To investigate whether the volume of white matter hyperintensity (WMH) extracted from FLAIR images changes when the imaging parameters of the original images are changed. METHODS: Seven healthy volunteers were imaged by changing the imaging parameter ETL of FLAIR images, and WMHs were extracted and their volumes were calculated by the automatic extraction software. The results were statistically analyzed to examine the relationship (Experiment 1). Simulated images with different SNRs were created by adding white noise to four examples of healthy volunteer images. The SNR of the simulated images simulated the SNR of the measured images of different ETLs. The WMH was extracted from the simulated images and its volume was calculated using the automatic extraction software (Experiment 2). RESULTS: Experiment 1 showed that there was no significant difference between FLAIR imaging parameters and WMH volume in automatic white matter signal analysis, except for some conditions. Experiment 2 showed that as the SNR of the original image decreased, the volume of high white matter signal extracted decreased. CONCLUSION: In automatic white matter signal analysis, WMH was shown to be small when the ETL of the FLAIR sequence was larger than normal and/or the SNR of the image was low.

Quality control system for mammographic breast positioning using deep learning.

This study proposes a deep convolutional neural network (DCNN) classification for the quality control and validation of breast positioning criteria in mammography. A total of 1631 mediolateral oblique mammographic views were collected from an open database. We designed two main steps for mammographic verification: automated detection of the positioning part and classification of three scales that determine the positioning quality using DCNNs. After acquiring labeled mammograms with three scales visually evaluated based on guidelines, the first step was automatically detecting the region of interest of the subject part by image processing. The next step was classifying mammographic positioning accuracy into three scales using four representative DCNNs. The experimental results showed that the DCNN model achieved the best positioning classification accuracy of 0.7836 using VGG16 in the inframammary fold and a classification accuracy of 0.7278 using Xception in the nipple profile. Furthermore, using the softmax function, the breast positioning criteria could be evaluated quantitatively by presenting the predicted value, which is the probability of determining positioning accuracy. The proposed method can be quantitatively evaluated without the need for an individual qualitative evaluation and has the potential to improve the quality control and validation of breast positioning criteria in mammography.

MRI safety for leave-on powdered hair thickeners under 1.5-T and 3.0-T MRI: measurement of deflection force, MRI artifact, and evaluation of preexamination screening.

This study aimed to clarify the magnetic resonance imaging (MRI) compatibility of leave-on powdered hair thickeners by evaluating the displacement force and image artifacts of commercially available leave-on powdered hair thickeners on MRI devices and their response to metal and ferromagnetic detectors. Thirteen types of leave-on powdered hair thickeners were studied: nine hair thickener and four foundation types. MRI systems of 1.5 T and 3.0 T were used. Deflection angles and MR image artifacts according to ASTM F2052 and F2119 were evaluated. Handheld metal and ferromagnetic detectors were used to investigate whether hair thickeners could be detected in screening before MRI examinations. The hair thickener type had a deflection angle of 0°, whereas the foundation type had a deflection angle of 90°, indicating a strong physical effect. Significant image artifacts appeared only on the foundation type. The foundation type reacted at distances of less than 10 cm only with a ferromagnetic detector. Foundation-type leave-on powdered hair thickeners containing magnetic substances exhibited strong physical effects and produced significant image artifacts, and those can only be detected by screening with a ferromagnetic detector.

Optimization of the fluid-attenuated inversion recovery (FLAIR) imaging for use in autopsy imaging of the brain region using synthetic MRI.

BACKGROUND: The failure of cerebrospinal fluid (CSF) signal suppression in postmortem fluid-attenuated inversion recovery (FLAIR) of the brain is a problem. OBJECTIVE: The present study was to clarify the relationship between the temperature of deceased persons and CSF T1, and to optimize the postmortem brain FLAIR imaging method using synthetic MRI. METHODS: Forehead temperature was measured in 15 deceased persons. Next, synthetic MRI of the brain was performed, the CSF T1 was measured, and the optimal TI was calculated. Two types of FLAIR images were obtained with the clinical and optimal TI. The relationship between forehead temperature and the CSF T1 and optimal TI was evaluated. The optimized FLAIR images were physically and visually evaluated. RESULTS: The CSF T1 and optimal TI were strongly correlated with forehead temperature. Comparing the average SNR and CNR ratios and visual evaluation scores of the two FLAIR images, those captured with the optimal TI showed statistically lower SNR, higher CNR, and higher visual evaluation scores (p< 0.01). CONCLUSIONS: Synthetic MRI enables the quantification of the CSF T1 resulting from postmortem temperature decreases and calculation of the optimal TI, which could aid in improving the failure of CSF signal suppression and in optimizing postmortem brain FLAIR imaging.

Incidence and risk factors of hepatocellular carcinoma in patients with hepatitis C who achieved a sustained virological response through direct-acting antiviral agents among the working population in Japan.

Background and Aim: The development of hepatocarcinogenesis after a sustained virological response (SVR) remains an important issue affecting the balance between treatment and occupational life of workers with chronic hepatitis C virus (HCV) infection in Japan. Here, we aimed to evaluate the hepatocellular carcinoma (HCC) reducing effect and risk factors for developing HCC after SVR in patients treated with direct-acting antiviral agents (DAAs) among the working population. Methods: We studied 2579 working patients with chronic HCV infection who achieved SVR after antiviral treatment. We compared the difference in the cumulative incidence of post-SVR HCC between the interferon (IFN)-based n = 1615 and DAA (n = 964) groups. The risk factors for post-SVR HCC development were determined in the DAA group. Results: After propensity score matching (n = 644 in each group), the HCC development rates were not significantly different between the groups (P = 0.186). Multivariate Cox regression and the cutoff values determined by the receiver operating characteristic curve analyses revealed that age ≥61 years, diabetes, lower serum albumin levels <4.0 g/dL at 24 weeks after the end of treatment (EOT), and higher serum α-fetoprotein levels ≥4.1 ng/mL at 24 weeks after the EOT were associated with the development of HCC. Conclusion: The HCC suppressing effect after SVR through DAA treatment is equivalent to that of IFN treatment in patients in the working population. Intensive follow-up is required after SVR with DAA treatment in Japanese workers with these risk factors to ensure the promotion of health and employment support.

Novel method for evaluating spatial resolution of magnetic resonance images.

Recently, several methods for evaluating the spatial resolution of magnetic resonance imaging have been reported. However, these methods are not simple and can only be used for specific devices. In this study, we develop a new method (the ladder method) and evaluate its measurement accuracy by adapting the International Electrotechnical Commission (IEC) method to evaluate the spatial resolution. First, the suitable condition for the ladder method was determined by numerical experiments. The ladder method uses a phantom with a periodic pattern which is based on IEC method. Subsequently, the ladder method is evaluated in terms of spatial resolution by dividing the standard deviation (SD) by the average signal in the region of interest (ROI) on the ladder phantom image. To evaluate the precision of the ladder method, it is compared with the modulation transfer function (MTF) calculated from an edge image. The numerical experiment result shows that the evaluation of the spatial resolution using the ladder method is viable, in which a single regression analysis's coefficient of correlation between ladder and MTF of 0.90 or higher is obtained for all evaluations. The ladder method can be assessed using only the signal mean value and SD in the ROI on the target image and exhibit a strong correlation with the MTF. Therefore, the ladder method is a promising method as a substitute for the MTF.

[Evaluation of Radiograph Accuracy in Skull X-ray Images Using Deep Learning].

PURPOSE: Accurate positioning is essential for radiography, and it is especially important to maintain image reproducibility in follow-up observations. The decision on re-taking radiographs is entrusting to the individual radiological technologist. The evaluation is a visual and qualitative evaluation and there are individual variations in the acceptance criteria. In this study, we propose a method of image evaluation using a deep convolutional neural network (DCNN) for skull X-ray images. METHOD: The radiographs were obtained from 5 skull phantoms and were classified by simple network and VGG16. The discrimination ability of DCNN was verified by recognizing the X-ray projection angle and the retake of the radiograph. DCNN architectures were used with the different input image sizes and were evaluated by 5-fold cross-validation and leave-one-out cross-validation. RESULT: Using the 5-fold cross-validation, the classification accuracy was 99.75% for the simple network and 80.00% for the VGG16 in small input image sizes, and when the input image size was general image size, simple network and VGG16 showed 79.58% and 80.00%, respectively. CONCLUSION: The experimental results showed that the combination between the small input image size, and the shallow DCNN architecture was suitable for the four-category classification in X-ray projection angles. The classification accuracy was up to 99.75%. The proposed method has the potential to automatically recognize the slight projection angles and the re-taking images to the acceptance criteria. It is considered that our proposed method can contribute to feedback for re-taking images and to reduce radiation dose due to individual subjectivity.

Development of an Eye-Tracking Image Manipulation System for Angiography: A Comparative Study.

RATIONALE AND OBJECTIVES: Appropriate image manipulation of angiographic image display systems during interventional radiology is performed by radiological technologists and/or nurses given instructions from radiologists. However, appropriate images might not be displayed because of communication errors. Therefore, we developed a manipulation system that uses an eye tracker. The study aimed to determine if an angiographic image display system can be manipulated as well by using an eye tracker as by using a mouse. MATERIALS AND METHODS: An angiographic image display system using an eye tracker to calculate the gaze position on the screen and state of fixation was developed. Fourteen radiological technologists participated in an observer study by manipulating 10 images for each of 5 typical cases frequently performed in angiography, such as renal tumor, cerebral aneurysm, liver tumor, uterine bleeding, and hypersplenism. We measured the time from the start to the end of manipulating a series of images required when using the eye tracker and the conventional mouse. In this study, the statistical processing was done using Excel and R and R studio. RESULTS: The average time required for all observers for completing all cases was significantly shorter when using the eye tracker than when using the mouse (10.4 ± 2.1 s and 16.9 ± 2.6 s, respectively; p< 0.001 by paired t test). CONCLUSION: Radiologists were able to manipulate an angiographic image display system directly by using the newly developed eye tracker system without touching contact devices, such as a mouse or angiography console. Therefore, communication error could be avoided.

Simultaneous brain structure segmentation in magnetic resonance images using deep convolutional neural networks.

In brain magnetic resonance imaging (MRI) examinations, rapidly acquired two-dimensional (2D) T1-weighted sagittal slices are typically used to confirm brainstem atrophy and the presence of signals in the posterior pituitary gland. Image segmentation is essential for the automatic evaluation of chronological changes in the brainstem and pituitary gland. Thus, the purpose of our study was to use deep learning to automatically segment internal organs (brainstem, corpus callosum, pituitary, cerebrum, and cerebellum) in midsagittal slices of 2D T1-weighted images. Deep learning for the automatic segmentation of seven regions in the images was accomplished using two different methods: patch-based segmentation and semantic segmentation. The networks used for patch-based segmentation were AlexNet, GoogLeNet, and ResNet50, whereas semantic segmentation was accomplished using SegNet, VGG16-weighted SegNet, and U-Net. The precision and Jaccard index were calculated, and the extraction accuracy of the six convolutional network (DCNN) systems was evaluated. The highest precision (0.974) was obtained with the VGG16-weighted SegNet, and the lowest precision (0.506) was obtained with ResNet50. Based on the data, calculation times, and Jaccard indices obtained in this study, segmentation on a 2D image may be considered a viable and effective approach. We found that the optimal automatic segmentation of organs (brainstem, corpus callosum, pituitary, cerebrum, and cerebellum) on brain sagittal T1-weighted images could be achieved using SegNet with VGG16.

[Creation of Brain Parenchymal Equivalent Phantom for Diffusion Weighted Image Using Microwave Oven: A Different Heating Method].

The acute micro cerebral infarction phantom was created to evaluate the detectability of acute micro cerebral infarction and optimize scan conditions in magnetic resonance imaging (MRI) examinations. The creation of the brain parenchyma phantom requires a hot stirrer that can be heated and stirred simultaneously. However, few hospitals and facilities have hot stirrers. The aim of our study was to use a microwave oven instead of a hot stirrer for the creation of a brain parenchyma phantom. Five phantoms using a hot stirrer and five phantoms using a microwave oven were created. The phantom creation time, T2 value, apparent diffusion coefficient (ADC) value, uniformity, and reproducibility of MR images were compared between the two creation methods. The phantom creation time, when using a microwave oven was 108±8 minutes, which was significantly shorter than that of 213± 48 minutes when using a hot stirrer. T2 values, ADC values, uniformity, and reproducibility were not significantly different. Therefore, it is easier to create an acute micro cerebral infarction phantom using a microwave oven compared to a hot stirrer.